Skip to content

Methodology

How we grade the evidence

Every study in our research library carries an evidence grade. It is a quick, honest signal of how much weight a finding can bear, so you can tell a strong human trial from an early animal experiment at a glance.

Independent by design

We take no commission on anything we cover. A study's grade reflects the strength of its evidence, never who funded it or whether it fits a product we sell.

The three grades

Strong

The finding rests on strong evidence in people. The effect has held up in rigorous studies, often more than one.

Typically: meta-analyses, large randomized controlled trials, or large long-term human cohorts.

Moderate

The finding is promising but not yet settled. The evidence points one way, but it is limited, indirect, or not yet repeated in people.

Typically: smaller randomized trials, solid observational studies, or well-run animal models.

Preliminary

An early signal worth watching. The idea is plausible and has some support, but it has not been shown to work in people yet.

Typically: cell and lab studies, case reports, or small early pilots.

What we weigh

Four things move a study up or down the scale.

01

Study design

Where the study sits on the evidence ladder below. Human trials outrank animal and lab work.

02

Sample size

How many participants took part, and for how long. Bigger, longer studies carry more weight.

03

Journal quality

Whether the work was peer-reviewed and published in a credible journal.

04

Reproducibility

Whether independent teams have found the same thing, or it still stands alone.

The evidence ladder

Not all evidence is equal. Findings near the top can bear real weight; findings near the bottom are clues, not conclusions.

StrongestWeakest
  1. 01Systematic reviews of human trialsStrongest: many randomized human studies, pooled and quality-checked
  2. 02A single human randomized trial (RCT)Randomization balances out confounders
  3. 03Human observational studiesUseful, but can be confounded and cannot prove cause
  4. 04Animal studies (mice, worms, flies)Great for hypotheses, often do not translate to people
  5. 05Cell and lab studiesA clue about mechanism, far from a human result
  6. 06'It makes biological sense'Plausibility is not proof of benefit or safety
  7. 07Expert opinionCan be informed, but it is not a measured effect
  8. 08Testimonials and anecdotesEasily fooled by placebo and selection
  9. 09Tradition and 'ancient wisdom'Weakest: no built-in protection against being wrong

A grade is a starting point

Evidence grades help you read the research quickly. They are not medical advice. Always talk to a qualified professional before changing your health routine.

Back to the research library